Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
1
Continuous process of carbon nanotubes synthesis by decomposition of methane using an arc-jet plasma
S. I. Choi*, J. S. Nam, J. I. Kim, T. H. Hwang, J. H. Seo, and S. H. Hong
Department of Nuclear Engineering, Seoul National University Seoul 151-742, Korea
Abstract
We present a method of producing carbon nanotubes by means of the thermal
plasma decomposition of methane (CH4) in an arc-jet plasma of high temperature
(5000–20000 K). Carbon nanotubes are produced under a floating condition by
introducing CH4 and a mixture of Ni-Y powders into the arc-jet plasma flame generated
by a non-transferred plasma torch. Material evaluations of the synthesized product by
transmission electron microscopy (TEM) and scanning electron microscopy (SEM)
reveals that the growth rate of carbon nanotubes is very high ,and that the produced
carbon nanotubes mainly have a multi-walled structure even though the single-walled
and coiled structures are rarely found. Since this process is continuously operable and
easily scalable, it is expected to be a promising technique for large-scale commercial
production of carbon nanotubes.
*Keywords
carbon nanotubes, arc-jet plasma, thermal plasma process, methane decomposition
2
1. Introduction Because of their unique properties, carbon nanotubes (CNTs) have a great potential
of practical applications to field emission display [1], reinforcing component [2],
hydrogen storage medium [3,4], and so on. However, in order to realize their industrial
applications, a continuous process for large-scale production of CNTs is essentially
required. To date, several methods such as arc discharge [5,6], laser ablation [7],
chemical vapor deposition (CVD) [8] and catalytic pyrolysis [9,10] have widely used
for CNTs production.
It is known that the conventional arc discharge method can generate highly
crystallized CNTs by high temperature ambience (above 5000 K) and achieve high
production rates of soot in which the yield of CNT is not satisfactory. But, this approach
has a drawback as an industrial technology due to its non-continuous process. The laser
ablation can easily produce the CNT of high purity. However, their production rates are
low, and the synthesis apparatus is expensive. The CVD method has merits to grow
aligned CNT and control diameter or length of CNT. Unfortunately, this approach has
difficulty in a continuous process because encapsulation of the metal particles, coated
on a substrate by diffusion of carbon, stops the growth process. In the catalytic pyrolysis
method, carbon nanotubes are produced in floating condition by introducing carbon and
catalyst sources simultaneously into an electric furnace. Therefore, this method is
recognized as a promising technique for large-scale production of industrial purpose due
to its continuous process and easy scale-up.
As a modified catalytic pyrolysis method, we present here a promising technique
for the continuous large-scale production of carbon nanotubes by decomposition of
hydrocarbon in an arc-jet plasma. The arc-jet plasma is generated by blowing a cold
3
plasma forming gas into the current path of dc arc discharge between two electrodes.
Since the arc-jet plasma has thermal plasma properties of high temperature (5000-20000
K) and high velocity (100-1000 m/s) [11], it has been widely used for plasma spray
coating [12,13], fine powder synthesis [14,15], and hazardous waste treatment by
pyrolysis of organic materials and melting of solid wastes [16]. When the arc-jet plasma
technique is applied to the CNT synthesis, this approach is analogous to the
conventional catalytic pyrolysis in respect that CNTs grow in a floating condition by
injecting hydrocarbon gas and catalyst powder simultaneously into a hot discharge
reactor zone. But, a thermal plasma reactor, instead of the electric furnace, is used as a
heat source for decomposition of hydrocarbon and catalyst. Therefore, the dc plasma
torch used in the reactor can provide an optimum synthesis ambience of very high
temperature, which is not achievable by the conventional electric furnace.
In the synthesis of CNTs by the arc-jet plasma method, its process features are
expected as follows: i) It is possible to achieve the high rates of hydrocarbon
decomposition and CNT growth resulted from the high temperature and enthalpy of
thermal plasma; ii) This technique itself is regarded as a process for the metal nano-
particle synthesis [15] due to evaporation of the metal powder of large size in the
thermal plasma of high temperature and enthalpy, and it does not use nano-sized metal
powders or organic metals as a catalyst source in contrast with the conventional
catalytic pyrolysis. The use of coarse powder is economically benefit for the catalyst
supply; iii) This new process can be easily scale-up through a plasma power-up.
Moreover, the plasma power-up can be achieved in a small volume without expansion
of the hot plasma zone; iv) The synthesized materials can be easily collected because of
4
one directional flow of the arc plasma jet, which permits a continuous synthesis process
with a continuous supply of carbon and catalyst sources.
In this paper, we present a successful synthesis of carbon nanotubes by
decomposition of methane (CH4) by the arc-jet thermal plasma generated by a non-
transferred plasma torch. The synthesized CNTs have been evaluated by the images of
high-resolution transmission electron microscopy (TEM) and scanning electron
microscopy (SEM). In our experimental study, we have found that the growth rate of
CNTs in the arc-jet plasma is very high, and that mainly multi-walled CNTs are
produced in high purity.
2. Experimental
A schematic diagram of the CNT synthesis system by hydrocarbon decomposition is
shown in Fig. 1. The arc-jet plasma is generated by a non-transferred plasma torch,
which consists of a tungsten cathode of conical shape and a copper anode of nozzle
shape. While a plasma gas is introduced into a channel between the cathode and the
nozzle anode, current paths form between the electrodes by dc arc discharge, and the
resultant thermal plasma jet is ejected from the anode nozzle exit. Both electrodes are
cooled by water in order to prevent them from erosion caused by enormous heat flux
from the arc.
In this experiment, a cathode diameter of 12 mm and a conical angle of 60◦, and an
inner anode diameter of 6 mm were taken. A mixture gas of argon and helium was used
as a plasma gas. Argon and helium flow rates were 10 slpm and 20 slpm, respectively.
Methane was used as a hydrocarbon, nickel and yttrium powders (Sigma-Aldrich) were
used as a metal precursor. CH4 of 1-4 slpm and a metal powder mixture of 0.2 g/min
5
(4:1 atomic ratio) were introduced, through a stainless tube of 1.5 mm in inner diameter,
into the arc-jet plasma near the anode nozzle exit. The mixture of nickel and yttrium
powders has been commonly used by other researchers for the conventional arc
discharge method [17,18]. The arc-jet plasma was generated with a dc current of 300 A,
and a loaded voltage of 48 V. The reactor pressure was maintained at 500 Torr.
The CNT samples produced in this experiment were collected by a stainless steel
collection tube for their material evaluation. They have been examined using high-
resolution TEM (JEOL JEM-3011), TEM (Philips CM-20), and field emission SEM
(Hitachi S-4700), which were operated at 300, 200, and 10 kV, respectively.
3. Results and Discussion
After the synthesis processes, some soot was obtained from the collection tubes,
and carbon nanotubes were found in the soot. All material analyses of the produced soot
were conducted without going through a purification process. Figure 2 shows the
macroscopic TEM images of the synthesized products of our experiment. As seen in a
very low-magnification TEM image of Fig. 2(a), most of the produced materials are
carbon nanotubes and a few amorphous carbons exist. The length of the CNTs is mainly
observed to be several μm, but long CNTs in a few tens μm are also found. It was
observed in our synthesis experiment that carbon vapors are condensed within 20 cm of
the plasma jet, of which velocity is generally known as higher than 100 m/s [11,15]. So,
the growth rate of the carbon nanotubes in the jet plasma is estimated in a range of 3-30
mm/min. The growth rate reported in a CVD method was suggested as the order of
μm/min [19]. The diffusion flame synthesis method reported a higher growth rate of
tens μm/min [20]. Therefore, the growth rate of CNTs produced by our arc-jet thermal
6
plasma method is much higher than that by other synthesis methods. An experimental
study on the CVD method has reported that the growth rate increases with temperature
[21]. Accordingly, it is inferred from this report that the much higher growth rates
obtained in our experiment is due to the very high temperature of the arc-jet thermal
plasma.
As observed in Fig. 2(b), outer diameters of the CNTs are widely distributed in a
range of 10 nm–50 nm and inner diameters are narrowly distributed around 10 nm. It
was suggested that outer diameter of CNT depends mainly on the catalyst size
evaporated during the synthesis process [22]. In the thermal plasma jet method, the
initially injected metal particles of large size are first evaporated in the hot region of
arc-jet thermal plasma and then condensed in the less hot region into nanoparticles,
which act as catalysts for forming CNTs. In a metal nanoparticle synthesis experiment
using an arc-jet plasma, metal particles of various sizes were synthesized by different
temperature distributions [15]. Therefore, the wide distribution of outer diameters of
CNTs obtained in our experiment appears to be caused by this general feature of arc-jet
plasma synthesis. By taking advantage of this feature of thermal plasma synthesis, it is
expected that the distribution of nanotubes diameters is controllable by regulating
operation conditions of the plasma torch for getting a desired temperature distribution.
In Fig. 3, the carbon nanotubes of multi-walled (a,b,c) and single-walled (c)
structures are observed in the soot produced by our jet plasma method. Figure 2(b) also
shows that most of produced carbon nanotubes are multi-walled ones. Assembles of
single-walled CNTs as seen in Fig. 3(c) are rarely found in the soot. The single-walled
CNTs layers appear to have large curvatures and relatively low crystallization degrees.
7
In contrast, the multi-walled CNT layers in Fig. 4(a), (b) show smaller curvature and
better crystallization degree compared to the single-walled CNTs.
Two different types of multi-walled CNTs are also observed in the TEM images in
the Fig. 3 : (a) bamboo-like nanotubes with some internal caps and (b) cylindrical
carbon nanotubes without internal caps. The appearance of the cylindrical CNTs
obtained by the thermal plasma method is similar to that by the conventional catalytic
pyrolysis method [23]. But the morphology of the bamboo-like nanotubes in our
experiment is different from that in the conventional catalytic pyrolysis or CVD method
[24,25]. Our CNTs have thick walls and small inner diameters, while those synthesized
by the pyrolysis or CVD have very thin walls and large diameters of 50-150 nm, while
ours have small diameters below 50 nm.
Coiled carbon nanotubes as a peculiar structure are rarely found in the soot as seen
in Fig. 3(d). as they have been sometimes produced by the CVD method [26,27]. The
coiled carbon nanotubes are believed to have exceptional properties, though their
applications have not been planed in an industrial scale yet [26]. The coiled carbon
nanotubes can be described by their coil diameter and pitch. In our experiment, the coil
diameter is about 100 nm and the pitch is about 50 nm. In the CVD method, coil
diameters are generally distributed in the range of 0.5-5 μm [28], and are much larger in
comparison with our method.
Figure 4 shows end tips of the produced CNTs.: (a) not encapsulated with catalytic
particle and (b) encapsulated with catalytic particle partly filled in their body.
Encapsulated or filled CNTs indicate that the injected large sized metal powders have
been sufficiently evaporated in the arc-jet thermal plasma of high temperature, and are
acted as a catalyst for the carbon nanotube growth.
8
Figure 5 shows the SEM image of the soot produced in our method reconfirms that
carbon nanotubes are synthesized in high purity as observed previously from the TEM
images of Fig. 2. This improved purity results of the arc-jet thermal plasma method
mainly comes from its unique property of very high temperature (above 5000 K), which
never be achievable by the conventional catalytic pyrolysis method.
The production rate of carbon soot yet seems low in our laboratory-scale
experiment. If CH4 of 1 slpm is fully extracted into solid carbon materials, the
production rate becomes about 0.54 g/min. Because the processing conditions of our
system are not fully optimized, the conversion rate of CH4 into carbon soot is below
10 %. But, our other study on a carbon black synthesis by this thermal plasma has
shown that the conversion rate of hydrocarbon into the soot could be increased by using
the optimization of processing conditions. In addition, even large amount of
hydrocarbons can be easily decomposed by a plasma power-up, and structures of
produced carbon nanotubes can be controlled by operation parameters.
4. Conclusion
We have demonstrated the synthesis method of carbon nanotubes by
decomposition of methane in the arc-jet thermal plasma. The SEM and TEM images
revealed that carbon nanotubes are successfully produced by this method. Especially, it
was found that the growth rate of carbon nanotubes is very high, and mainly multi-
walled CNTs in high purity and a few other structures of nanotubes were produced.
However, the conversion rate of CH4 into carbon soot and the degree of crystallization
of carbon nanotubes was somewhat low.
9
The thermal plasma process is not yet optimized and further studies on production
of carbon nanotubes by this method are required. But, we feel that this method is a
promising technique for the large-scale commercial production of carbon nanotubes
because of its numerous advantages of continuous operation, high growth rate of carbon
nanotubes, easy scale-up of plasma power, and easy collection of products.
Acknowledgment
This work was supported by Korea Institute of Science and Technology Evaluation
and Planning (KISTEP) of the Ministry of Science and Technology in Korea.
10
References
[1] Q.H. Wang, A.A. Setlur, J.M. Lauerhaas, J.Y. Dai, E.W. Seelig and R.P.H. Chang,
Appl. Phys. Lett. 72 (1998) 2912
[2] D. Qian, E.C. Dickey, R. Andrews and T. Rantell, Appl. Phys. Lett. 76 (2000) 2868
[3] C. Liu, Y.Y. Fan, M. Liu, H.T. Cong, H.M. Cheng and M.S. Dresselhaus, Science
286 (1999) 1127
[4] Anyuan Cao, Hongwei Zhu, Xianfeng Zhang, Xuesong Li, Dianbo Ruan, Cailu Xu,
Binqing Wei, Ji Liang and Dehai Wu, Chem. Phys. Lett. 342 (2001) 510
[5] T.W. Ebbensen and P.M. Ajayan, Nature 358 (1992) 220
[6] S. Iijima and T. Ichihashi, Nature 363 (1993) 603
[7] Andreas Thess, Roland Lee, Pavel Nikolaev, Hongjie Dai, Pierre Petit, Jerome
Robert, Chunhui Xu, Young Hee Lee, Seong Gon Kim, Andrew G. Rinzler, Daniel T.
Colbert, Gustavo E. Scuseria, David Tomanek, John E. Fischer and Richard E. Smalley,
Science 273 (1996) 483
[8] Young Chul Choi, Young Min Shin, Seong Chu Lim, Dong Jae Bae, Young Hee
Lee, Byung Soo Lee, Dong-Chul Chung, J. Appl. Phys. 88 (2000) 4898
[9] H.M. Cheng, F.Li, L.L. He, X. Sun and M.S. Dresselhaus, Appl. Phys. Lett. 72
(1998) 3282
[10] Lijie Ci, Jinquan Wei, Bingqing Wei, Ji Liang, Cailu Xu and Dehai Wu, Carbon 39
(2001) 329
[11] Maher I. Boulos, IEEE T. Plasma Sci. 19 (1991) 1078
[12] Pierre Fauchais and Armelle Vardelle, IEEE T. Plasma Sci. 25 (1997) 1258
[13] M. Vardelle, A. Vardelle, A.C. Leger, P. Fauchais, and D. Gobin, J. Therm. Spray
Tech. 4 (1995) 50
11
[14] P. Fauchais, A. Vardelle and A. Denoirjean, Surf. Coat. Tech. 97 (1997) 66
[15] N. Rao, S. Girshick, J. Heberlein, P. McMurry, S. Jones, D. Hansen and B. Micheel,
Plasma Chem. Plasma Process. 15 (1995) 581
[16] H.R. Snyder, C.B. Fleddermann and J.M. Gahl, Waste Manage. 16 (1996) 289
[17] Yoshinori Ando, Xiuluo Zhao, Kaori Hirahara, Kazutomo Suenaga, Shunji
Bandow and Sumio Iijima, Chem. Phys. Lett. 323 (2000) 580
[18] Zujin Shi, Yongfu Lian, Fu Hui Liao, Xihuang Zhou, Zhennan Gu, Y. Zhang, S.
Iijima, Hongdong Li, Kwok To Yue and Shu-Lin Zhang, J. Phys. Chem. Solids 61
(2000) 1031
[19] Z.F. Ren, Z.P. Huang, J.W.Xu, J.H. Wang, P. Bush, M.P. Siegal, P.N. Provencion,
Science 282 (1998) 1105
[20] Randall L. Vander Wal, Thomas M. Ticich and Valerie E. Curtis, Chem. Phys. Lett.
323 (2000) 217
[21] Cheol Jin Lee, Jeunghee Park, Yoon Huh and Jeong Yong Lee, Chem. Phys. Lett.
343 (2001) 33
[22] W.Z. Li, D.Z. Wang, S.X. Yang, J.G. Wen and Z.F. Ren, Chem. Phys. Lett. 335
(2001) 141
[23] Yun Tack Lee, Nam Seo Kim, Jeunghee Park, Hae Beom Han, Young Sang Choi,
Hyun Ryu and hwack Joo Lee, Chem. Phys. Lett. 372 (2003) 853
[24] Xianbao Wang, Wenping Hu, Yunqu Liu, chenfeng Long, Yu Xu, Shuqin Zhou,
Daoben Zhu and Liming Dai, Carbon 39 (2001) 1533
[25] Cheol Jin Lee and Jeunghee Park, Carbon 39 (2001) 1891
[26] Seiji Motojima and Quiqin Chen, J. Appl. Phys. 85 (1999) 3919
[27] J.B. Bai, Mater. Lett. 57 (2003) 2629
12
[28] Federic Fabry, Gilles Flamant and Laurent Fulcheri, Chem. Eng. Sci. 56 (2001)
2123
13
Figure Captions
Fig. 1
Schematic diagram of an arc-jet thermal plasma reactor for carbon nanotubes synthesis
by decomposition of methane
Fig. 2
Low-magnification TEM images of the morphology of as-grown carbon nanotubes
show (a) length distributions of the CNTs and (b) diameter distributions of the
synthesized CNTs
Fig. 3
TEM images show the carbon nanotubes of different structures produced by the arc-jet
thermal plasma method: (a) bamboo-like multi-walled CNT (b) multi-walled CNT (c)
single-walled CNTs, and (d) coiled CNTs
Fig. 4
High-resolution TEM images of as-grown carbon nanotubes: (a) open-tip CNTs (b)
encapsulated tip and partially filled CNTs (c) wall structure of a cylindrical multi-
walled CNT, and (d) wall structure of a bamboo-like multi-walled CNT
Fig. 5
SEM image of the carbon nanotubes obtained from the collection tubes
14
Fig. 1
15
Fig. 2
16
Fig. 3
17
Fig. 4
18
Fig. 5